US10019178B2 - Method and related device for determining management mode of shared virtual memory page - Google Patents

Method and related device for determining management mode of shared virtual memory page Download PDF

Info

Publication number
US10019178B2
US10019178B2 US14/749,869 US201514749869A US10019178B2 US 10019178 B2 US10019178 B2 US 10019178B2 US 201514749869 A US201514749869 A US 201514749869A US 10019178 B2 US10019178 B2 US 10019178B2
Authority
US
United States
Prior art keywords
memory page
mode
shared virtual
virtual memory
access
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US14/749,869
Other languages
English (en)
Other versions
US20150293720A1 (en
Inventor
KingTin LAM
Wangbin Zhu
Youliang Yan
Cho-Li Wang
Jinghao Shi
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Assigned to HUAWEI TECHNOLOGIES CO., LTD. reassignment HUAWEI TECHNOLOGIES CO., LTD. ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: LAM, Kingtin, WANG, CHO-LI, YAN, YOULIANG, ZHU, WANGBIN, Shi, Jinghao
Publication of US20150293720A1 publication Critical patent/US20150293720A1/en
Application granted granted Critical
Publication of US10019178B2 publication Critical patent/US10019178B2/en
Active legal-status Critical Current
Adjusted expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F15/00—Digital computers in general; Data processing equipment in general
    • G06F15/16—Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs
    • G06F15/163—Interprocessor communication
    • G06F15/167—Interprocessor communication using a common memory, e.g. mailbox
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0806—Multiuser, multiprocessor or multiprocessing cache systems
    • G06F12/0815—Cache consistency protocols
    • G06F12/0837—Cache consistency protocols with software control, e.g. non-cacheable data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0806—Multiuser, multiprocessor or multiprocessing cache systems
    • G06F12/0815—Cache consistency protocols
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/0802—Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
    • G06F12/0806—Multiuser, multiprocessor or multiprocessing cache systems
    • G06F12/0842—Multiuser, multiprocessor or multiprocessing cache systems for multiprocessing or multitasking
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/10—Address translation
    • G06F12/109—Address translation for multiple virtual address spaces, e.g. segmentation
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
    • G06F3/0614—Improving the reliability of storage systems
    • G06F3/0619—Improving the reliability of storage systems in relation to data integrity, e.g. data losses, bit errors
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0646—Horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
    • G06F3/065—Replication mechanisms
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
    • G06F3/0653—Monitoring storage devices or systems
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
    • G06F3/0601—Interfaces specially adapted for storage systems
    • G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
    • G06F3/067—Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/10—Address translation
    • G06F12/1072—Decentralised address translation, e.g. in distributed shared memory systems
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/10—Providing a specific technical effect
    • G06F2212/1016—Performance improvement
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/10—Providing a specific technical effect
    • G06F2212/1016—Performance improvement
    • G06F2212/1024—Latency reduction
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/60—Details of cache memory
    • G06F2212/6042—Allocation of cache space to multiple users or processors
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2212/00—Indexing scheme relating to accessing, addressing or allocation within memory systems or architectures
    • G06F2212/65—Details of virtual memory and virtual address translation
    • G06F2212/656—Address space sharing

Definitions

  • the present disclosure relates to the field of computer technologies, and in particular to a method and a related device for determining a management mode of a shared virtual memory page.
  • a multi-core/many-core processor scenario needs to allow for cache coherence, which may be solved by a software cache coherence mechanism or a hardware cache coherence mechanism.
  • the software cache coherence mechanism primarily relates to two possible modes: a distributed shared memory (DSM) mode and a shared physical memory (SPM) mode.
  • DSM distributed shared memory
  • SPM shared physical memory
  • the DSM mode/SPM mode may be applied in a con-cache-coherent (NCC) many-core system to solve the cache coherence issue.
  • NCC con-cache-coherent
  • SMC Software Managed Cache Coherence
  • Embodiments of the present disclosure provide a method and a related device for determining a management mode of a shared virtual memory page, with a view to enhancing performance of accessing a shared virtual memory.
  • the present disclosure provides a method for determining a management mode of a shared virtual memory page, that include monitoring access characteristics of access performed by a process for the shared virtual memory page and changing the management mode of the shared virtual memory page to a shared physical memory mode if the monitored access characteristics meet a first set condition and a current management mode of the shared virtual memory page is a distributed shared memory mode.
  • the method further may include changing the management mode of the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics meet a second set condition and the current management mode of the shared virtual memory page is a shared physical memory mode.
  • the changing the management mode of the shared virtual memory page to a distributed shared memory mode includes in a first physical memory space, allocating a first physical memory page to the shared virtual memory page, copying data stored on a second physical memory page to the first physical memory page, where the second physical memory page belongs to a shared physical memory space, and the second physical memory page is a physical memory page currently mapped from the shared virtual memory page, creating a mapping relationship between the shared virtual memory page and the first physical memory page; and recording the distributed shared memory mode as the management mode of the shared virtual memory page.
  • the first set condition includes that an access frequency is less than or equal to a first threshold, an access mode is read-write access or write access, a frequency of write access is greater than a third threshold, or a percentage of write access is greater than a fourth threshold.
  • the second set condition includes that the access frequency is greater than a second threshold, the access mode is read-only access, a frequency of read access is greater than a fifth threshold, or a percentage of read access is greater than a sixth threshold, where the second threshold is greater than or equal to the first threshold.
  • the changing the management mode of the shared virtual memory page to a shared physical memory mode includes creating a mapping relationship between the shared virtual memory page and a third physical memory page, where the third physical memory page belongs to the shared physical memory space; and recording the shared physical memory mode as the management mode of the shared virtual memory page, where data stored on the physical memory page previously mapped from the shared virtual memory page remains unchanged.
  • the changing the management mode of the shared virtual memory page to a shared physical memory mode includes synchronizing changed data in data stored on a fourth physical memory page to a third physical memory page; creating a mapping relationship between the shared virtual memory page and the third physical memory page, where the third physical memory page belongs to the shared physical memory space; and recording the shared physical memory mode as the management mode of the shared virtual memory page, where the fourth physical memory page is a physical memory page previously mapped from the shared virtual memory page.
  • the changing the management mode of the shared virtual memory page to a shared physical memory mode if the monitored access characteristics meet a first set condition and the current management mode of the shared virtual memory page is a distributed shared memory mode includes using an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored for each of the multiple processes that access the shared virtual memory page, as the monitored access characteristics of the multiple processes that access the shared virtual memory page; or, using monitored access characteristics of one of the multiple processes that access the shared virtual memory page as the monitored access characteristics of the multiple processes that access the shared virtual memory page; and changing the management mode corresponding to the multiple processes on the shared virtual memory page to a shared physical memory mode if the monitored access characteristics of access performed by the multiple processes for the shared virtual memory page meet the first condition and if the current management mode corresponding to the multiple processes on the shared virtual memory page is a distributed shared memory mode.
  • the changing the management mode of the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics meet a second set condition and the current management mode of the shared virtual memory page is a shared physical memory mode includes using an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored for each of the multiple processes that access the shared virtual memory page, as the monitored access characteristics of the multiple processes that access the shared virtual memory page; or, using monitored access characteristics of one of the multiple processes that access the shared virtual memory page as the monitored access characteristics of the multiple processes that access the shared virtual memory page; and changing the management mode corresponding to the multiple processes on the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics of access performed by the multiple processes for the shared virtual memory page meet the second condition and if the current management mode corresponding to the multiple processes on the shared virtual memory page is a shared physical memory mode.
  • the present disclosure provides a method for determining a management mode of a shared virtual memory page, including monitoring access characteristics of access performed by a process for the shared virtual memory page and changing the management mode of the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics meet a second set condition and the current management mode of the shared virtual memory page is a shared physical memory mode.
  • the changing the management mode of the shared virtual memory page to a distributed shared memory mode includes in a first physical memory space, allocating a first physical memory page to the shared virtual memory page copying data stored on a second physical memory page to the first physical memory page, where the second physical memory page belongs to a shared physical memory space, and the second physical memory page is a physical memory page currently mapped from the shared virtual memory page, creating a mapping relationship between the shared virtual memory page and the first physical memory page, and recording the distributed shared memory mode as the management mode of the shared virtual memory page.
  • the changing the management mode of the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics meet a second set condition and the current management mode of the shared virtual memory page is a shared physical memory mode includes using an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored for each of the multiple processes that access the shared virtual memory page, as the monitored access characteristics of the multiple processes that access the shared virtual memory page; or, using monitored access characteristics of one of the multiple processes that access the shared virtual memory page as the monitored access characteristics of the multiple processes that access the shared virtual memory page; and changing the management mode corresponding to the multiple processes on the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics of access performed by the multiple processes for the shared virtual memory page meet the second condition and if the current management mode corresponding to the multiple processes on the shared virtual memory page is a shared physical memory mode.
  • the second set condition includes an access frequency is greater than a second threshold, an access mode is read-only access, a frequency of read access is greater than a fifth threshold, or a percentage of read access is greater than a sixth threshold.
  • the present disclosure provides an apparatus for determining a management mode of a shared virtual memory page, a monitoring unit, configured to monitor access characteristics of access performed by a process for the shared virtual memory page and a first remapping unit connected to the monitoring unit, configured to change the management mode of the shared virtual memory page to a shared physical memory mode if the access characteristics monitored by the monitoring unit meet a first set condition and a current management mode of the shared virtual memory page is a distributed shared memory mode.
  • the apparatus further includes a second remapping unit connected to the monitoring unit, configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the access characteristics monitored by the monitoring unit meet a second set condition and a current management mode of the shared virtual memory page is a shared physical memory mode.
  • the second remapping unit is specifically configured to: if the access characteristics monitored by the monitoring unit meet a second set condition and the current management mode of the shared virtual memory page is a shared physical memory mode, then in a first physical memory space, allocate a first physical memory page to the shared virtual memory page; copy data stored on a second physical memory page to the first physical memory page, where the second physical memory page belongs to a shared physical memory space, and the second physical memory page is a physical memory page currently mapped from the shared virtual memory page; create a mapping relationship between the shared virtual memory page and the first physical memory page; and record the distributed shared memory mode as the management mode of the shared virtual memory page.
  • the first possible implementation manner of the third aspect, or the second possible implementation manner of the third aspect in a third possible implementation manner the first set condition includes that an access frequency is less than or equal to a first threshold, an access mode is read-write access or write access, a frequency of write access is greater than a third threshold, or a percentage of write access is greater than a fourth threshold the monitoring unit is specifically configured to monitor an access frequency of access performed by the process for the shared virtual memory page, where the first remapping unit is specifically configured to change the management mode of the shared virtual memory page to a shared physical memory mode if the access frequency of access performed by the process for the shared virtual memory page, which is monitored by the monitoring unit, is less than or equal to the first threshold, and if the current management mode of the shared virtual memory page is a distributed shared memory mode or, the monitoring unit is specifically configured to monitor an access mode of access performed by the process for the shared virtual memory page the first remapping unit is specifically configured to change the management mode of the shared virtual memory page to a
  • the monitoring unit is specifically configured to monitor the frequency of write access performed by the process for the shared virtual memory page, where the first remapping unit is specifically configured to change the management mode of the shared virtual memory page to a shared physical memory mode if the frequency of write access performed by the process for the shared virtual memory page, which is monitored by the monitoring unit, is greater than the third threshold, and if the current management mode of the shared virtual memory page is a distributed shared memory mode or the monitoring unit is specifically configured to monitor a percentage of write access performed by the process for the shared virtual memory page, where the first remapping unit is specifically configured to change the management mode of the shared virtual memory page to a shared physical memory mode if the percentage of write access performed by the process for the shared virtual memory page, which is monitored by the monitoring unit, is greater than the fourth threshold, and if the current management mode of the shared virtual memory page is a distributed shared memory mode.
  • the second set condition includes that the access frequency is greater than a second threshold, the access mode is read-only access, a frequency of read access is greater than a fifth threshold, or a percentage of read access is greater than a sixth threshold, where the second threshold is greater than or equal to the first threshold, where the monitoring unit is specifically configured to monitor an access frequency of access performed by the process for the shared virtual memory page, where the second remapping unit is specifically configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the access frequency monitored by the monitoring unit is greater than the second threshold and the current management mode of the shared virtual memory page is a shared physical memory mode or the monitoring unit is specifically configured to monitor an access mode of access performed by the process for the shared virtual memory page the second remapping unit is specifically configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the access mode monitored by the monitoring unit is read-only access and the current management mode of the shared
  • the first remapping unit is specifically configured to: if the access characteristics monitored by the monitoring unit meet a first set condition and the current management mode of the shared virtual memory page is a distributed shared memory mode, create a mapping relationship between the shared virtual memory page and a third physical memory page, where the third physical memory page belongs to the shared physical memory space; and record the shared physical memory mode as the management mode of the shared virtual memory page, where data stored on the physical memory page previously mapped from the shared virtual memory page remains unchanged.
  • the first remapping unit is specifically configured to: synchronize changed data in data stored on a fourth physical memory page to a third physical memory page if the access characteristics monitored by the monitoring unit meet a first set condition and the current management mode of the shared virtual memory page is a distributed shared memory mode; create a mapping relationship between the shared virtual memory page and the third physical memory page, where the third physical memory page belongs to the shared physical memory space; and record the shared physical memory mode as the management mode of the shared virtual memory page, where the fourth physical memory page is a physical memory page previously mapped from the shared virtual memory page.
  • the first remapping unit is specifically configured to: use an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored by the monitoring unit for each of the multiple processes that access the shared virtual memory page, as the monitored access characteristics of the multiple processes that access the shared virtual memory page; or, use the access characteristics, which are monitored by the monitoring unit for one of the multiple processes that access the shared virtual memory page, as the monitored access characteristics of the multiple processes that access the shared virtual memory page; and change the management mode corresponding to the multiple processes on the shared virtual memory page to a shared physical memory mode if the monitored access characteristics of access performed by the processes for the shared virtual memory page meet the first condition and if the current management mode corresponding to the multiple processes on the shared virtual memory page is a distributed shared memory mode.
  • the second remapping unit is specifically configured to: use an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored by the monitoring unit for each of the multiple processes that access the shared virtual memory page, as the monitored access characteristics of the multiple processes that access the shared virtual memory page; or, use the access characteristics, which are monitored by the monitoring unit for one of the multiple processes that access the shared virtual memory page, as the monitored access characteristics of the multiple processes that access the shared virtual memory page; and change the management mode corresponding to the multiple processes on the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics of access performed by the multiple processes for the shared virtual memory page meet the second condition and if the current management mode corresponding to the multiple processes on the shared virtual memory page is a shared physical memory mode.
  • the present disclosure provides an apparatus for determining a management mode of a shared virtual memory page, which may include: a monitoring unit, configured to monitor access characteristics of access performed by a process for the shared virtual memory page; and a second remapping unit connected to the monitoring unit, configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the access characteristics monitored by the monitoring unit meet a second set condition and a current management mode of the shared virtual memory page is a shared physical memory mode.
  • the second remapping unit is specifically configured to: if the access characteristics monitored by the monitoring unit meet a second set condition and the current management mode of the shared virtual memory page is a shared physical memory mode, then in a first physical memory space, allocate a first physical memory page to the shared virtual memory page; copy data stored on a second physical memory page to the first physical memory page, where the second physical memory page belongs to a shared physical memory space, and the second physical memory page is a physical memory page currently mapped from the shared virtual memory page; create a mapping relationship between the shared virtual memory page and the first physical memory page; and record the distributed shared memory mode as the management mode of the shared virtual memory page.
  • the second set condition includes that an access frequency is greater than a second threshold, an access mode is read-only access, a frequency of read access is greater than a fifth threshold, or a percentage of read access is greater than a sixth threshold
  • the monitoring unit is specifically configured to monitor an access frequency of access performed by the process for the shared virtual memory page, where the second remapping unit is specifically configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the access frequency monitored by the monitoring unit is greater than the second threshold and the current management mode of the shared virtual memory page is a shared physical memory mode; or, the monitoring unit is specifically configured to monitor an access mode of access performed by the process for the shared virtual memory page, the second remapping unit is specifically configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the access mode monitored by the monitoring unit is read-only access and the current management mode of the shared virtual memory page is a shared
  • the monitoring unit may be specifically configured to monitor a percentage of read access performed by the process for the shared virtual memory page, where the second remapping unit is specifically configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the percentage of read access monitored by the monitoring unit is greater than the sixth threshold and the current management mode of the shared virtual memory page is a shared physical memory mode.
  • the second remapping unit is specifically configured to: use an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored by the monitoring unit for each of the multiple processes that access the shared virtual memory page, as the monitored access characteristics of the multiple processes that access the shared virtual memory page; or, use the access characteristics, which are monitored by the monitoring unit for one of the multiple processes that access the shared virtual memory page, as the monitored access characteristics of the multiple processes that access the shared virtual memory page; and change the management mode corresponding to the multiple processes on the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics of access performed by the multiple processes for the shared virtual memory page meet the second condition and if the current management mode corresponding to the multiple processes on the shared virtual memory page is a shared physical memory mode.
  • the present disclosure provides a computer system, including: a bus, a processor connected to the bus, and a memory connected to the bus, where the processor uses the bus to call codes in the memory to monitor access characteristics of access performed by a process for the shared virtual memory page; and change the management mode of the shared virtual memory page to a shared physical memory mode if the monitored access characteristics meet a first set condition and a current management mode of the shared virtual memory page is a distributed shared memory mode.
  • the present disclosure provides a computer system, includes a bus, a processor connected to the bus, and a memory connected to the bus, where the processor uses the bus to call codes in the memory to monitor access characteristics of access performed by a process for the shared virtual memory page; and change the management mode of the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics meet a second set condition and a current management mode of the shared virtual memory page is a shared physical memory mode.
  • the access characteristics of the shared virtual memory page are monitored; and the management mode of the shared virtual memory page is changed to a shared physical memory mode if the monitored access characteristics meet the first set condition and the current management mode of the shared virtual memory page is a distributed shared memory mode; or, the management mode of the shared virtual memory page is changed to a distributed shared memory mode if the monitored access characteristics meet the second set condition and the current management mode of the shared virtual memory page is a shared physical memory mode.
  • the mechanism of resetting the management mode of the shared virtual memory page according to a monitoring result of the access characteristics achieves better matching between the management mode and the access characteristics, brings the advantages of the DSM mode or the SPM mode into better play, and lays a mechanism foundation for enhancing performance of accessing the shared virtual memory.
  • FIG. 1 is a schematic flowchart of a method for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure
  • FIG. 2 - a is a schematic diagram of a shared virtual address space in DSM mode according to an embodiment of the present disclosure
  • FIG. 2 - b is a schematic diagram of a private physical memory space in DSM mode according to an embodiment of the present disclosure
  • FIG. 3 is a schematic flowchart of another method for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure
  • FIG. 4 is a schematic flowchart of another method for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure
  • FIG. 5 - a is a schematic diagram of an apparatus for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure
  • FIG. 5 - b is a schematic diagram of an apparatus for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure
  • FIG. 5 - c is a schematic diagram of an apparatus for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure
  • FIG. 6 is a schematic structural diagram of a computer system according to the present disclosure.
  • FIG. 7 is a schematic diagram of another computer system according to an embodiment of the present disclosure.
  • Embodiments of the present disclosure provide a method and a related device for determining a management mode of a shared virtual memory page, with a view to enhancing performance of accessing a shared virtual memory.
  • a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units enumerated expressly, but may include other steps or units that are not enumerated expressly or are inherent in the process, method, product, or device.
  • a method for determining a management mode of a shared virtual memory page may include: monitoring access characteristics of access performed by a process for the shared virtual memory page; and changing the management mode of the shared virtual memory page to an SPM mode if the monitored access characteristics meet a first set condition and a current management mode of the shared virtual memory page is a DSM mode, and/or, changing the management mode of the shared virtual memory page to a DSM mode if the monitored access characteristics meet a second set condition and the current management mode of the shared virtual memory page is an SPM mode.
  • FIG. 1 is a schematic flowchart of a method for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure. As shown in FIG. 1 , a method for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure may include the following content.
  • Block 101 discloses the monitoring of access characteristics of access performed by a process for the shared virtual memory page.
  • the access characteristics of access performed by N processes for the shared virtual memory page can be monitored, where N is a positive integer.
  • N is a positive integer.
  • the N process may refer to any process currently existent in the system or a specific process; if N is greater than 1, the N processes may refer to part or all of the processes currently existent in the system.
  • the N processes may refer to part or all of the processes, which currently exist in the system and will possibly access the first shared virtual memory page.
  • a performance counter or other hardware or programs of the CPU may be used to monitor the access characteristics of the shared virtual memory page.
  • various access characteristics of access performed by the process for the shared virtual memory page may be monitored.
  • the access characteristics may include access frequency and/or access mode, and so on.
  • the access modes may include read-only access, read-write access (the read-write access refers to access that involves both read and write), write access, and so on. That is, the access characteristics such as access frequency and/or access mode of access performed by the process for the shared virtual memory page can be monitored.
  • the access frequency of access performed by the process for the first shared virtual memory page may refer to the times the process accesses the first shared virtual memory page in a unit time.
  • the access frequency of access performed by a process for the first shared virtual memory page may refer to the times the process accesses the first shared virtual memory page in a unit time.
  • the access frequency of access performed by multiple processes for the first shared virtual memory page may refer to the times the multiple processes access the first shared virtual memory page in a unit time.
  • the access frequency of access performed by the process for the first shared virtual memory page may refer to a percentage of the number of times of access performed by the process for the first shared virtual memory page to the total number of times of access performed by the process for N 1 shared virtual memory pages in a duration (such as a unit time), where the N 1 shared virtual memory pages include the first shared virtual memory page and N 1 is a positive integer greater than 1.
  • the way of calculating the access frequency is not limited to the above example, and other ways of calculation compliant with the laws of the nature may be defined as required.
  • Block 102 discloses changing the management mode of the first shared virtual memory page to an SPM mode if the monitored access characteristics meet the first set condition and the current management mode of the first shared virtual memory page is a DSM mode, and/or, change the management mode of the first shared virtual memory page to a DSM mode if the monitored access characteristics meet the second set condition and the current management mode of the first shared virtual memory page is an SPM mode.
  • the first set condition and the second set condition may vary with the monitored access characteristics. For example, if the monitored access characteristics are an access frequency, the first set condition and the second set condition may include set conditions related to the access frequency; if the monitored access characteristics are an access mode, the first set condition and the second set condition may include set conditions related to the access mode; and, if the monitored access characteristics include the access frequency and the access mode, the first set condition and the second set condition may include set conditions related to the both access frequency and the access mode.
  • the first set condition may include that the access frequency is less than or equal to a first threshold (the first threshold may be 1000 times per second, 2000 times per second, 5000 times per second, or another frequency value); and the second set condition may include that the access frequency is greater than a second threshold, where the first threshold may be less than or equal to the second threshold, and the second threshold may be 1000 times per second, 2000 times per second, 3000 times per second, 5000 times per second, or another frequency value, where the second threshold is greater than or equal to the first threshold.
  • a first threshold may be 1000 times per second, 2000 times per second, 5000 times per second, or another frequency value
  • the management mode of the first shared virtual memory page may be changed to an SPM mode if the monitored access frequency is less than or equal to the first threshold and the current management mode of the first shared virtual memory page is a DSM mode, and/or, the management mode of the first shared virtual memory page may be changed to a DSM mode if the monitored access frequency is greater than the second threshold and the current management mode of the first shared virtual memory page is an SPM mode.
  • Such a policy may be regarded as a consumptive selection policy according to access data.
  • the first set condition may include that the access mode is write access or read-write access (or the percentage of the write access exceeds a fourth threshold or the frequency of the write access exceeds a third threshold), where the scenario of existence of only write access may be regarded as a fact that the percentage of the number of times of the write access to the total number of times of the read-write access is 100%; and the second set condition may include that the percentage of the read-only access or the read access exceeds a sixth threshold or that the frequency of the read access exceeds a fifth threshold, where the sum of the fourth threshold and the sixth threshold may be greater than or equal to 1.
  • the management mode of the first shared virtual memory page may be changed to an SPM mode if the monitored access mode is write access or read-write access (or the percentage of the write access exceeds a fourth threshold or the frequency of the write access exceeds a third threshold) and the current management mode of the first shared virtual memory page is a DSM mode; and/or, the management mode of the first shared virtual memory page may be changed to a DSM mode if the monitored access mode is read-only access (or the percentage of the read access exceeds the sixth threshold or the frequency of the read access exceeds the fifth threshold) and the current management mode of the first shared virtual memory page is an SPM mode, where the read-only access scenario may be regarded as a fact that the percentage of the number of times of the read access to the total number of times of the read-write access is 100%.
  • the percentage of the write access refers to a percentage of the number of times of the write access to the total number of times of the read-write access
  • the percentage of the read access refers to a percentage of the number of times of the read access to the total number of times of the read-write access.
  • Other circumstances can be inferred similarly. It can be understood that such a policy may be regarded as a consumptive selection policy according to access synchronization.
  • the management mode of the first shared virtual memory page may be changed to a DSM mode in different ways.
  • the changing the management mode of the first shared virtual memory page to a DSM mode may include: in a first physical memory space, allocating a first physical memory page to the first shared virtual memory page; copying data stored on a second physical memory page to the first physical memory page, where the second physical memory page belongs to a shared physical memory space (the shared physical memory space is a physical memory space that can be shared by several cores), and the second physical memory page is a physical memory page currently mapped from the first shared virtual memory page; creating a mapping relationship between the first shared virtual memory page and the first physical memory page; and recording the DSM mode as the management mode of the first shared virtual memory page.
  • the changing the management mode of the shared virtual memory page to an SPM mode may include: creating a mapping relationship between the first shared virtual memory page and the third physical memory page, where the third physical memory page belongs to the shared physical memory space (the shared physical memory space is a physical memory space that can be shared by several cores); and recording the SPM mode as the management mode of the first shared virtual memory page, where the data stored on the physical memory page previously mapped from the first shared virtual memory page remains unchanged.
  • the changing the management mode of the shared virtual memory page to an SPM mode may include: synchronizing changed data in the data stored on a fourth physical memory page to a third physical memory page; creating a mapping relationship between the first shared virtual memory page and the third physical memory page, where the third physical memory page belongs to the shared physical memory space (the shared physical memory space is a physical memory space that can be shared by several cores); and recording the SPM mode as the management mode of the first shared virtual memory page, where the fourth physical memory page is a physical memory page previously mapped from the first shared virtual memory page.
  • the changed data in the data stored on the fourth physical memory page may refer to a value-varying variable stored on the fourth physical memory page.
  • a variable y 1 , a variable y 2 , and a variable y 3 stored on the third physical memory page are copied to the fourth physical memory page, after the variable y 1 , the variable y 2 , and the variable y 3 are stored on the fourth physical memory page, the values of the variable y 1 and the variable y 2 remain unchanged, but the value of the variable y 2 changes from 0 to 1 as against the original copy, and therefore, the changed data in the data stored on the fourth physical memory page include the variable y 2 , where the variable y 2 is the changed data, and other circumstances can be inferred similarly.
  • the DSM mode/SPM mode may be recorded as the management mode of the first shared virtual memory page in a variety of ways.
  • a management mode flag bit is set for the first shared virtual memory page, and the management mode flag bit may be set to different values to identify whether the management mode of the first shared virtual memory page is a DSM mode or an SPM mode.
  • the management mode of the first shared virtual memory page which is recorded in the page information structure body of a process (or multiple processes or all processes), may be recorded as a DSM mode or an SPM mode.
  • Other ways of recording are not enumerated here exhaustively.
  • the management mode (such as DSM mode or SPM mode) may be set for each shared virtual memory page (such as the first shared virtual memory page). In this scenario, for all the processes that access the first shared virtual memory page, the management mode of the first shared virtual memory page is the same.
  • the management mode (such as DSM mode or SPM mode) corresponding to each process may also be set for each shared virtual memory page (such as the first shared virtual memory page). In this scenario, for all the processes that access the first shared virtual memory page, the management mode of the first shared virtual memory page is the same or different. For example, assuming that process A and process B exist, the management mode corresponding to process A on the first shared virtual memory page may be a DSM mode, and the management mode corresponding to process B on the first shared virtual memory page may be an SPM mode.
  • the access characteristics of access performed by the first process for the first shared virtual memory page may be monitored, where the first process may include M 1 processes and M 1 is a positive integer; and the management mode corresponding to the first process on the first shared virtual memory page may be changed to an SPM mode if the monitored access characteristics meet the first set condition and the current management mode corresponding to the first process on the first shared virtual memory page is a DSM mode; and/or, the management mode corresponding to the first process on the first shared virtual memory page may be changed to a DSM mode if the monitored access characteristics meet the second set condition and the current management mode corresponding to the first process on the first shared virtual memory page is an SPM mode.
  • the management modes corresponding to the multiple processes on the first shared virtual memory page are the same (DSM mode or SPM mode).
  • the access characteristics of access performed by the second process for the first shared virtual memory page may be monitored, where the second process includes M 2 processes, M 2 is a positive integer, and the second process is different from the first process; the management mode corresponding to the second process on the first shared virtual memory page may be changed to an SPM mode if the monitored access characteristics of access performed by the second process for the first shared virtual memory page meet the third set condition and the current management mode corresponding to the second process on the first shared virtual memory page is a DSM mode; and/or, the management mode corresponding to the second process on the first shared virtual memory page may be changed to a DSM mode if the monitored access characteristics of access performed by the second process for the first shared virtual memory page meet the fourth set condition and the current management mode corresponding to the second process on the first shared virtual memory page is an SPM mode, where the third set condition is the same as or different from the first condition, and the fourth set condition is the same as or different from the second condition.
  • the management mode corresponding to the first process on the first shared virtual memory page may be the same as or different from the management mode corresponding to the second process on the first shared virtual memory page, that is, the management modes corresponding to different processes on the same shared virtual memory page may be the same or different, which is called an asynchronous mapping mode.
  • the management mode of the first shared virtual memory page is determined directly by monitoring the access characteristics of access performed by the first process for the first shared virtual memory page, it is not necessary to implement the step of monitoring the access characteristics of access performed by the second process for the first shared virtual memory page, but the current management mode corresponding to the second process on the first shared virtual memory page may be directly set to the management mode that is the same as the current management mode corresponding to the first process on the first shared virtual memory page.
  • the management mode corresponding to the first process on the first shared virtual memory page is a DSM mode according to a result of monitoring the access characteristics of access performed by the first process for the first shared virtual memory page
  • the management mode corresponding to the second process on the first shared virtual memory page may also be set to a DSM mode, and other circumstances can be inferred similarly.
  • the monitored access characteristics of one of the multiple processes may be used as the monitored access characteristics of access performed by the multiple processes for the first shared virtual memory page.
  • the monitored integrated access characteristics of access performed by the multiple processes for the first shared virtual memory page may be used as the monitored access characteristics of access performed by the multiple processes for the first shared virtual memory page.
  • an average, a weighted average, a sum, or a weighted sum of the access characteristics may be used as the monitored access characteristics of access performed by the multiple processes for the first shared virtual memory page.
  • an average, a weighted average, a sum, or a weighted sum of the access frequency (or the frequency of read access or the percentage of read access or the frequency of write access or the percentage of write access), which is monitored respectively for each of the multiple processes that access the first shared virtual memory page (the weighting value corresponding to each process may be set as required, and may surely be all 1s), may be used as the monitored access characteristics of access performed by the multiple processes for the first shared virtual memory page.
  • the monitored access characteristics of access performed by the first process or the second process for the first shared virtual memory page may be determined in the above exemplary manner. Other circumstances can be inferred similarly.
  • all processes of a concurrent program may have the same shared virtual address space.
  • Each process maps part or all of the virtual memory pages in the shared virtual address space to a private physical memory space maintained by this process.
  • each process sees a complete shared virtual address space but cannot perceive that the shared data included in a virtual memory page in the shared virtual address space is actually a private physical memory space maintained by other processes.
  • Each process may perform any data operation for the shared virtual address space.
  • a DSM bottom layer synchronizes data between all processes by using an on-chip network or a system shared physical memory accessible by all processes. Multiple processes of one concurrent program may run on a processor core, or each process runs on a separate processor core.
  • region a in FIG. 2 - a is a shared virtual address space of programs; four processes P 0 ⁇ P 3 run on different processor cores; they have a uniform shared virtual address space for storing shared data, where the first region in the virtual address space is mapped to the private physical memory space of P 0 , the second region is mapped to the private physical memory space of P 1 , the third region is mapped to the private physical memory space of P 2 , and the fourth region is mapped to the private physical memory space of P 3 .
  • a home region in FIG. 2 - b is a private physical address space of each process, and each process maintains a non-overlapped physical memory space (the part indicated by “home” in FIG. 2 - b ) mapped from the shared virtual address space.
  • the DSM protocol copies the physical memory page inclusive of variable z, which is maintained by process 2 , to the physical memory of process 1 (see the part indicated by “cache” of P 1 in FIG. 2 - b ). In this way, process 1 can access variable z.
  • the DSM protocol maintains coherence of variable z in the system to ensure that process 1 and process 2 see the same variable z.
  • the DSM mode maps the virtual memory page in the shared virtual address space to the private physical memory space maintained by the process, and the private physical memory space data is generally configured as being cached by L1 Cache and L2 Cache. Therefore, when the CPU reads data, the cache hit rate (cache hits) is very high; and, when the CPU writes data, the cache hit rate is very high too, which is very favorable to program performance.
  • the shared virtual address space is mapped to the shared physical address space in all processes of the same concurrent program to achieve software cache coherence.
  • the SPM mode generally disables L2 Cache, and L1 Cache is generally set to employ a write through (WT) policy.
  • WT write through
  • a memory page generally maintains only one copy in the shared physical memory space, which saves the physical memory space.
  • the DSM mode and the SPM mode have their respective advantages and disadvantages, and the DSM mode and the SPM mode have a certain preference for different access characteristics (for example, different access frequencies, different access modes). Therefore, in the solution in the embodiment of the present disclosure, the access characteristics of access performed by the process for the shared virtual memory page are monitored, and the monitored access characteristics are used as a basis for determining whether to reset the management mode of the shared virtual memory page.
  • Such an adaptive mechanism between the access characteristics and the management mode is good for enhancing performance of accessing the shared virtual memory page and therefore good for improving performance of the corresponding program and the user experience.
  • the access characteristics of the shared virtual memory page are monitored; and the management mode of the shared virtual memory page is changed to an SPM mode if the monitored access characteristics meet the first set condition and the current management mode of the shared virtual memory page is a DSM mode, and/or, the management mode of the shared virtual memory page is changed to a DSM mode if the monitored access characteristics meet the second set condition and the current management mode of the shared virtual memory page is an SPM mode.
  • the mechanism of resetting the management mode of the shared virtual memory page according to a monitoring result of the access characteristics achieves better matching between the management mode and the access characteristics, brings the advantages of the DSM mode or the SPM mode into better play, and lays a mechanism foundation for enhancing performance of accessing the shared virtual memory.
  • FIG. 3 is a schematic flowchart of another method for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure.
  • the other method for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure may include the following content:
  • Block 301 discloses monitoring access characteristics of access performed by N processes for the shared virtual memory page X.
  • N is a positive integer greater than 1.
  • the N processes may refer to part or all of the processes currently existent in the system.
  • the N processes may refer to part or all of the processes, which currently exist in the system and will possibly access the shared virtual memory page X.
  • the shared virtual memory page X in this embodiment may correspond to the first shared virtual memory page in the preceding embodiment.
  • various access characteristics of access performed by N processes for the shared virtual memory page X may be monitored, and an example given in this embodiment is to detect the access frequency of access performed by N processes for the shared virtual memory page X.
  • the access frequency of access performed by the N processes for the shared virtual memory page X may refer to the times the processes access the shared virtual memory page X in a unit time. In other embodiments of the present disclosure, the access frequency of access performed by the N processes for the shared virtual memory page X may refer to a percentage of the number of times of access performed by the N processes for the shared virtual memory page X to the total number of times of access performed by the N processes for N 1 shared virtual memory pages in a duration (such as a unit time), where the N 1 shared virtual memory pages include the shared virtual memory page X and N 1 is a positive integer greater than 1.
  • the way of calculating the access frequency of access performed by the N processes for the shared virtual memory page X is not limited to the above example, and other ways of calculation compliant with the laws of the nature may be defined as required.
  • Block 302 discloses judging whether the monitored access frequency of access performed by the N processes for the shared virtual memory page X is less than or equal to a first threshold, and if yes, perform the elements in Block 303 or, if no, perform step 304 .
  • Block 303 discloses that if the monitored access frequency of access performed by the N processes for the shared virtual memory page X is less than or equal to the first threshold and the current management mode corresponding to the N processes on the shared virtual memory page X is a DSM mode, change the management mode corresponding to the N processes on the shared virtual memory page X to an SPM mode, and return to step 301 ; and, if the current management mode corresponding to the N processes on the shared virtual memory page X is an SPM mode, return to step 301 .
  • Block 304 discloses that if the monitored access frequency of access performed by the N processes for the shared virtual memory page X is greater than a second threshold and the current management mode corresponding to the N processes on the shared virtual memory page X is an SPM mode, change the management mode corresponding to the N processes on the shared virtual memory page X to a DSM mode, and return to step 301 ; and, if the current management mode corresponding to the N processes on the shared virtual memory page X is a DSM mode, return to step 301 .
  • the first threshold may be 1000 times per second, 2000 times per second, 5000 times per second, or another frequency value; and the first threshold may be less than or equal to the second threshold, and the second threshold may be 1000 times per second, 2000 times per second, 3000 times per second, 5000 times per second, or another frequency value.
  • the first threshold may be 20%, 25%, 30%, 35% or another percentage, the first threshold may be less than or equal to the second threshold, and the second threshold may be 20%, 25%, 30%, 35% or another percentage. It can be understood that such a policy may be regarded as a consumptive selection policy according to access data.
  • management mode corresponding to the N processes on the shared virtual memory page X may be changed to a DSM mode in different ways.
  • the changing the management mode corresponding to the N processes on the shared virtual memory page X to a DSM mode may include: in a first physical memory space, allocating a first physical memory page to the first shared virtual memory page; copying data stored on a second physical memory page to the first physical memory page, where the second physical memory page belongs to a shared physical memory space, and the second physical memory page is a physical memory page currently mapped from the shared virtual memory page X; creating a mapping relationship between the shared virtual memory page X and the first physical memory page; and recording the DSM mode as the management mode corresponding to the N processes on the shared virtual memory page X.
  • the management mode of the shared virtual memory page X which is recorded in the page information structure body of the N processes, may be all recorded as a DSM mode.
  • the changing the management mode corresponding to the N processes on the shared virtual memory page X to an SPM mode may include: creating a mapping relationship between the shared virtual memory page X and a third physical memory page, where the third physical memory page belongs to the shared physical memory space; and recording the SPM mode as the management mode corresponding to the N processes on the shared virtual memory page X.
  • the management mode of the shared virtual memory page X which is recorded in the page information structure body of the N processes, may be all recorded as an SPM mode.
  • the method may further include: synchronizing changed data in the data stored on a fourth physical memory page previously mapped from the shared virtual memory page X to a third physical memory page.
  • the data synchronization step may be omitted.
  • the fourth physical memory page is a physical memory page previously mapped from the shared virtual memory page X.
  • the DSM mode and the SPM mode have a certain preference for different access frequencies. Therefore, in the solution in this embodiment, the access characteristics of access performed by the N processes for the shared virtual memory page are monitored, and the monitored access characteristics are used as a basis for determining whether to reset the management mode corresponding to the N processes on the shared virtual memory page.
  • Such an adaptive mechanism between the access characteristics and the management mode is good for enhancing performance of accessing the shared virtual memory page and therefore good for improving performance of the corresponding program and the user experience.
  • FIG. 4 is a schematic flowchart of another method for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure.
  • the other method for determining a management mode of a shared virtual memory page according to an embodiment of the present disclosure may include the following content:
  • Block 401 it is monitoring access characteristics of access performed by process A and process B for the shared virtual memory page X is shown.
  • various access characteristics of access performed by process A and process B for the shared virtual memory page X may be monitored, and an example given in this embodiment is to detect the access frequency of access performed by process A and process B for the shared virtual memory page X.
  • the shared virtual memory page X in this embodiment may correspond to the first shared virtual memory page in the preceding embodiment.
  • the frequency of access performed by process A and process B for the shared virtual memory page X may refer to the times process A and process B access the shared virtual memory page X in a unit time.
  • the frequency of access performed by process A and process B for the shared virtual memory page X may also refer to a percentage of the number of times of process A and process B accessing the shared virtual memory page X to the total number of times of process A and process B accessing N 1 shared virtual memory pages in a duration (such as a unit time), where the N 1 shared virtual memory pages include the shared virtual memory page X, and N 1 is a positive integer greater than 1.
  • block 402 judging whether the monitored frequency of access performed by process A for the shared virtual memory page X is less than or equal to a first threshold is disclosed, If the result of block 402 is if yes, then the elements in block 403 are performed, and if the result of block 402 is no, then the elements in block 404 are performed.
  • Block 403 discloses that if the monitored frequency of access performed by process A for the shared virtual memory page X is less than or equal to the first threshold and the current management mode corresponding to process A on the shared virtual memory page X is a DSM mode, change the management mode corresponding to process A on the shared virtual memory page X to an SPM mode, and return to step 401 ; and, if the current management mode corresponding to process A on the shared virtual memory page X is an SPM mode, return to step 401 .
  • Block 404 discloses that if the monitored frequency of access performed by process A for the shared virtual memory page X is greater than a second threshold and the current management mode corresponding to process A on the shared virtual memory page X is an SPM mode, change the management mode corresponding to process A on the shared virtual memory page X to a DSM mode, and return to step 401 ; and, if the current management mode corresponding to process A on the shared virtual memory page X is a DSM mode, return to step 401 .
  • Block 405 discloses judging whether the monitored frequency of access performed by process B for the shared virtual memory page X is less than or equal to a first threshold.
  • Block 406 discloses that if the monitored frequency of access performed by process B for the shared virtual memory page X is less than or equal to the first threshold and the current management mode corresponding to process B on the shared virtual memory page X is a DSM mode, change the management mode corresponding to process B on the shared virtual memory page X to an SPM mode, and return to step 401 ; and, if the current management mode corresponding to process B on the shared virtual memory page X is an SPM mode, return to step 401 .
  • Block 407 discloses that if the monitored frequency of access performed by process B for the shared virtual memory page X is greater than a second threshold and the current management mode corresponding to process B on the shared virtual memory page X is an SPM mode, change the management mode corresponding to process B on the shared virtual memory page X to a DSM mode, and return to step 401 ; and, if the current management mode corresponding to process B on the shared virtual memory page X is a DSM mode, return to step 401 .
  • steps 402 ⁇ 404 and steps 405 ⁇ 407 do not necessarily occur in a definite order, and may be performed synchronously or consecutively.
  • the first threshold may be 1000 times per second, 2000 times per second, 5000 times per second, or another frequency value; and the first threshold may be less than or equal to the second threshold, and the second threshold may be 1000 times per second, 2000 times per second, 3000 times per second, 5000 times per second, or another frequency value.
  • the first threshold may be 20%, 25%, 30%, 35% or another percentage, the first threshold may be less than or equal to the second threshold, and the second threshold may be 20%, 25%, 30%, 35% or another percentage.
  • Such a policy may be regarded as a consumptive selection policy according to access data.
  • management mode corresponding to process A (or process B) of the shared virtual memory page X may be changed to a DSM mode in different ways.
  • the changing the management mode corresponding to process A (or process B) of the first shared virtual memory page to a DSM mode may include: in a first physical memory space, allocating a first physical memory page to the shared virtual memory page X; copying data stored on a second physical memory page to the first physical memory page, where the second physical memory page belongs to a shared physical memory space, and the second physical memory page is a physical memory page currently mapped from the shared virtual memory page X; creating a mapping relationship between the shared virtual memory page X and the first physical memory page; and recording the DSM mode as the management mode corresponding to process A (or process B) of the shared virtual memory page X.
  • the management mode of the shared virtual memory page X which is recorded in the page information structure body of process A (or process B), may be all recorded as a DSM mode.
  • the second physical memory page may remain effective subsequently, and the second physical memory page may serve as a synchronization channel between processes that use the DSM mode as a management mode of the shared virtual memory page X, or between processes that use the SPM mode as a management mode of the shared virtual memory page X, or between a process that uses the DSM mode as a management mode of the shared virtual memory page X and a process that uses the SPM mode as a management mode of the shared virtual memory page X, where the synchronization channel is specific to the shared virtual memory page X.
  • the changing the management mode corresponding to process A (or process B) of the shared virtual memory page X to an SPM mode may include: creating a mapping relationship between the shared virtual memory page X and a third physical memory page, where the third physical memory page belongs to the shared physical memory space; and recording the SPM mode as the management mode corresponding to process A (or process B) of the shared virtual memory page X.
  • the management mode of the shared virtual memory page X which is recorded in the page information structure body of process A (or process B), may be all recorded as an SPM mode.
  • the method may further include: synchronizing changed data in the data stored on a fourth physical memory page previously mapped from the shared virtual memory page X to a third physical memory page.
  • the data synchronization step may be omitted.
  • the DSM mode and the SPM mode have a certain preference for different access frequencies. Therefore, in the solution in this embodiment, the access characteristics of access performed by process A and process B for the shared virtual memory page are monitored respectively, and the monitored access characteristics are used as a basis for determining respectively whether to reset the management mode corresponding to process A and process B on the shared virtual memory page.
  • Such an adaptive mechanism between the access characteristics and the management mode is good for enhancing performance of accessing the shared virtual memory page and therefore good for improving performance of the corresponding program and the user experience.
  • the access characteristics of different processes serve as a basis for determining whether to reset the management mode corresponding to each process on the shared virtual memory page, which makes the management mode more adaptive and enhances access performance.
  • an apparatus 500 for determining a management mode of a shared virtual memory page may include a monitoring unit 510 and a first remapping unit 520 .
  • the monitoring unit 510 is configured to monitor access characteristics of access performed by a process for the shared virtual memory page.
  • the first remapping unit 520 connected to the monitoring unit 510 is configured to change the management mode of the shared virtual memory page to an SPM mode if the access characteristics monitored by the monitoring unit 510 meet a first set condition and a current management mode of the shared virtual memory page is a DSM mode.
  • the apparatus 500 for determining a management mode of a shared virtual memory page may further include a second remapping unit 530 connected to the monitoring unit 510 , configured to change the management mode of the shared virtual memory page to a DSM mode if the access characteristics monitored by the monitoring unit 510 meet a second set condition and a current management mode of the shared virtual memory page is an SPM mode.
  • a second remapping unit 530 connected to the monitoring unit 510 , configured to change the management mode of the shared virtual memory page to a DSM mode if the access characteristics monitored by the monitoring unit 510 meet a second set condition and a current management mode of the shared virtual memory page is an SPM mode.
  • the apparatus 500 for determining a management mode of a shared virtual memory page may include a second remapping unit 530 connected to the monitoring unit 510 , configured to change the management mode of the shared virtual memory page to a DSM mode if the access characteristics monitored by the monitoring unit 510 meet a second set condition and a current management mode of the shared virtual memory page is an SPM mode.
  • the first set condition and the second set condition may vary with the monitored access characteristics. For example, if the monitored access characteristics are an access frequency, the first set condition and the second set condition may include set conditions related to the access frequency; if the monitored access characteristics are an access mode, the first set condition and the second set condition may include set conditions related to the access mode; if the monitored access characteristics include the access frequency and the access mode, the first set condition and the second set condition may include set conditions related to the both access frequency and the access mode.
  • the first set condition may include that the access frequency is less than or equal to a first threshold (for example, the first threshold may be 1000 times per second, 2000 times per second, 5000 times per second, or another frequency value); and the second set condition may include that the access frequency is greater than a second threshold, where the first threshold may be less than or equal to the second threshold, and the second threshold may be 1000 times per second, 2000 times per second, 3000 times per second, 5000 times per second, or another frequency value, where the second threshold is greater than or equal to the first threshold.
  • a first threshold for example, the first threshold may be 1000 times per second, 2000 times per second, 5000 times per second, or another frequency value
  • the second threshold may be 1000 times per second, 2000 times per second, 3000 times per second, 5000 times per second, or another frequency value, where the second threshold is greater than or equal to the first threshold.
  • the management mode of the first shared virtual memory page may be changed to an SPM mode if the monitored access frequency is less than or equal to the first threshold and the current management mode of the first shared virtual memory page is a DSM mode, and/or, the management mode of the first shared virtual memory page may be changed to a DSM mode if the monitored access frequency is greater than the second threshold and the current management mode of the first shared virtual memory page is an SPM mode.
  • a policy may be regarded as a consumptive selection policy according to access data.
  • the first set condition may include that the access mode is write access or read-write access (or the percentage of the write access exceeds a fourth threshold or the frequency of the write access exceeds a third threshold); and the second set condition may include that the percentage of the read-only access or read access exceeds a sixth threshold or that the frequency of the read access exceeds a fifth threshold, where the sum of the fourth threshold and the sixth threshold may be greater than or equal to 1.
  • the management mode of the first shared virtual memory page may be changed to an SPM mode if the monitored access mode is write access or read-write access (or the percentage of the write access exceeds a fourth threshold or the frequency of the write access exceeds a third threshold) and the current management mode of the first shared virtual memory page is a DSM mode; and/or, the management mode of the first shared virtual memory page may be changed to a DSM mode if the monitored access mode is read-only access (or the percentage of the read access exceeds the sixth threshold or the frequency of the read access exceeds the fifth threshold) and the current management mode of the first shared virtual memory page is an SPM mode, and other circumstances may be inferred similarly. It can be understood that such a policy may be regarded as a consumptive selection policy according to access synchronization.
  • the first set condition may include that the access frequency is less than or equal to a first threshold, or the access mode is read-write access or write access, or the frequency of the write access is greater than a third threshold, or the percentage of the write access is greater than a fourth threshold.
  • the monitoring unit 510 is specifically configured to monitor the access frequency of access performed by the process for the shared virtual memory page.
  • the first remapping unit 520 may be specifically configured to change the management mode of the shared virtual memory page to a shared physical memory mode if the access frequency monitored by the monitoring unit 510 is less than or equal to the first threshold and the current management mode of the shared virtual memory page is a distributed shared memory mode.
  • the monitoring unit 510 is specifically configured to monitor the access mode of access performed by the process for the shared virtual memory page; and the first remapping unit 520 may be specifically configured to change the management mode of the shared virtual memory page to a shared physical memory mode if the access mode monitored by the monitoring unit 510 is read-write access or write access and if the current management mode of the shared virtual memory page is a distributed shared memory mode.
  • the monitoring unit 510 is specifically configured to monitor the frequency of write access performed by the process for the shared virtual memory page; and the first remapping unit 520 may be specifically configured to change the management mode of the shared virtual memory page to a shared physical memory mode if the frequency of write access monitored by the monitoring unit 510 is greater than a third threshold and if the current management mode of the shared virtual memory page is a distributed shared memory mode.
  • the monitoring unit 510 is specifically configured to monitor the percentage of access performed by the process for the shared virtual memory page; and the first remapping unit 520 may be specifically configured to change the management mode of the shared virtual memory page to a shared physical memory mode if the percentage of write access monitored by the monitoring unit 510 is greater than a fourth threshold and if the current management mode of the shared virtual memory page is a distributed shared memory mode.
  • the second set condition includes that the access frequency is greater than a second threshold, the access mode is read-only access, the frequency of the read access is greater than a fifth threshold, or the percentage of the read access is greater than a sixth threshold, where the second threshold may be greater than or equal to the first threshold.
  • the monitoring unit 510 is specifically configured to monitor the frequency of access performed by the process for the shared virtual memory page; and the second remapping unit 530 may be specifically configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the frequency of access monitored by the monitoring unit 510 is greater than a second threshold and if the current management mode of the shared virtual memory page is a shared physical memory mode.
  • the monitoring unit 510 is specifically configured to monitor the mode of access performed by the process for the shared virtual memory page; and the second remapping unit 530 may be specifically configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the mode of access monitored by the monitoring unit 510 is read-only access and if the current management mode of the shared virtual memory page is a shared physical memory mode.
  • the monitoring unit 510 is specifically configured to monitor the frequency of read access performed by the process for the shared virtual memory page; and the second remapping unit 530 may be specifically configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the frequency of read access monitored by the monitoring unit 510 is greater than a fifth threshold and if the current management mode of the shared virtual memory page is a shared physical memory mode.
  • the monitoring unit 510 is specifically configured to monitor the percentage of read access performed by the process for the shared virtual memory page; and the second remapping unit 530 may be specifically configured to change the management mode of the shared virtual memory page to a distributed shared memory mode if the percentage of read access monitored by the monitoring unit 510 is greater than a sixth threshold and if the current management mode of the shared virtual memory page is a shared physical memory mode.
  • the first remapping unit 520 may be specifically configured to: if the access characteristics monitored by the monitoring unit 510 meet a first set condition and the current management mode of the shared virtual memory page is a DSM mode, create a mapping relationship between the shared virtual memory page and a third physical memory page, where the third physical memory page belongs to the shared physical memory space; and record the SPM mode as the management mode of the shared virtual memory page, where data stored on the physical memory page previously mapped from the shared virtual memory page remains unchanged.
  • the first remapping unit 520 may also be specifically configured to: synchronize changed data in data stored on a fourth physical memory page to a third physical memory page if the access characteristics monitored by the monitoring unit 510 meet a first set condition and the current management mode of the shared virtual memory page is a DSM mode; create a mapping relationship between the shared virtual memory page and the third physical memory page, where the third physical memory page belongs to the shared physical memory space; and record the SPM mode as the management mode of the shared virtual memory page, where the fourth physical memory page is a physical memory page previously mapped from the shared virtual memory page and the data stored on the physical memory page is changed.
  • the second remapping unit 530 may be specifically configured to: if the access characteristics monitored by the monitoring unit 510 meet a second set condition and the current management mode of the shared virtual memory page is an SPM mode, then in a first physical memory space, allocate a first physical memory page to the shared virtual memory page; copy data stored on a second physical memory page to the first physical memory page, where the second physical memory page belongs to a shared physical memory space, and the second physical memory page is a physical memory page currently mapped from the shared virtual memory page; create a mapping relationship between the shared virtual memory page and the first physical memory page; and record the DSM mode as the management mode of the shared virtual memory page.
  • the monitoring unit 510 is specifically configured to monitor the access characteristics of access performed by the first process for the shared virtual memory page, where the first process includes M 1 processes and M 1 is a positive integer.
  • the first remapping unit 520 may be specifically configured to change the management mode corresponding to the first process on the shared virtual memory page to an SPM mode if the access characteristics of access performed by the first process for the shared virtual memory page, which are monitored by the monitoring unit 510 , meet the first set condition, and, if the current management mode corresponding to the first process on the shared virtual memory page is a DSM mode.
  • the second remapping unit 530 may be specifically configured to change the management mode corresponding to the first process on the shared virtual memory page to a DSM mode if the access characteristics of access performed by the first process for the shared virtual memory page, which are monitored by the monitoring unit 510 , meet the second set condition, and, if the current management mode corresponding to the first process on the shared virtual memory page is an SPM mode.
  • the monitored access characteristics of access performed by one of the M 1 processes for the shared virtual memory page may be used as the monitored access characteristics of access performed by the first process for the shared virtual memory page.
  • the monitored integrated access characteristics of access performed by the M 1 processes for the first shared virtual memory page may be used as the monitored access characteristics of access performed by the first process for the shared virtual memory page.
  • an average, a weighted average, a sum, or a weighted sum of the access characteristics, which are monitored respectively for each of the M 1 processes that access the shared virtual memory page may be used as the monitored access characteristics of access performed by the first process for the shared virtual memory page.
  • the monitoring unit 510 is further configured to monitor the access characteristics of access performed by the second process for the shared virtual memory page, where the second process includes M 2 processes, M 2 is a positive integer, and the second process is different from the first process.
  • the first remapping unit 520 may be further configured to change the management mode corresponding to the second process on the shared virtual memory page to an SPM mode if the access characteristics of access performed by the second process for the shared virtual memory page, which are monitored by the monitoring unit 510 , meet the first set condition, and, if the current management mode corresponding to the second process on the shared virtual memory page is a DSM mode.
  • the second remapping unit 530 is configured to change the management mode corresponding to the second process on the shared virtual memory page to a DSM mode if the access characteristics of access performed by the second process for the shared virtual memory page, which are monitored by the monitoring unit 510 , meet the second set condition, and, if the current management mode corresponding to the second process on the shared virtual memory page is an SPM mode.
  • the monitored access characteristics of access performed by one of the M 2 processes for the shared virtual memory page may be used as the monitored access characteristics of access performed by the second process for the shared virtual memory page.
  • the monitored integrated access characteristics of access performed by the M 2 processes for the first shared virtual memory page may be used as the monitored access characteristics of access performed by the second process for the shared virtual memory page.
  • an average, a weighted average, a sum, or a weighted sum of the access characteristics, which are monitored respectively for each of the M 2 processes that access the shared virtual memory page may be used as the monitored access characteristics of access performed by the second process for the shared virtual memory page.
  • the first remapping unit 520 may be specifically configured to: use an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored by the monitoring unit 510 for each process in the first process (or the second process) that accesses the shared virtual memory page, as the monitored access characteristics of the first process (or the second process) that accesses the shared virtual memory page; or, use the access characteristics, which are monitored by the monitoring unit 510 for one process in the first process (or the second process) that accesses the shared virtual memory page, as the monitored access characteristics of the first process (or the second process) that accesses the shared virtual memory page; and change the management mode corresponding to the first process (or the second process) on the shared virtual memory page to a shared physical memory mode if the monitored access characteristics of access performed by the processes for the shared virtual memory page meet the first condition and if the current management mode corresponding to the first process (or the second process) on the
  • the second remapping unit 530 may be specifically configured to: use an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored by the monitoring unit 510 for each process in the first process (or the second process) that accesses the shared virtual memory page, as the monitored access characteristics of the first process (or the second process) that accesses the shared virtual memory page; or, use the access characteristics, which are monitored by the monitoring unit 510 for one process in the first process (or the second process) that accesses the shared virtual memory page, as the monitored access characteristics of the first process (or the second process) that accesses the shared virtual memory page; and change the management mode corresponding to the first process (or the second process) on the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics of access performed by the processes for the shared virtual memory page meet the second condition and if the current management mode corresponding to the first process (or the second process) on
  • the apparatus 500 for determining a management mode of a shared virtual memory page may be deployed on a communication device such as an access network device or a core network device or a router or another communication device, and the apparatus 500 for determining a management mode of a shared virtual memory page may also be deployed on a server or a personal computer or another computer device.
  • a communication device such as an access network device or a core network device or a router or another communication device
  • the apparatus 500 for determining a management mode of a shared virtual memory page may also be deployed on a server or a personal computer or another computer device.
  • function modules in the apparatus 500 for determining a management mode of a shared virtual memory page in this embodiment may be implemented according to the method described in the method embodiment.
  • functions of function modules in the apparatus 500 for determining a management mode of a shared virtual memory page in this embodiment may be implemented according to the method described in the method embodiment.
  • the apparatus 500 for determining a management mode of a shared virtual memory page monitors the access characteristics of the shared virtual memory page; and changes the management mode of the shared virtual memory page to a shared physical memory mode if the monitored access characteristics meet the first set condition and the current management mode of the shared virtual memory page is a distributed shared memory mode, or, changes the management mode of the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics meet the second set condition and the current management mode of the shared virtual memory page is a shared physical memory mode.
  • the mechanism of resetting the management mode of the shared virtual memory page according to a monitoring result of the access characteristics achieves better matching between the management mode and the access characteristics, brings the advantages of the DSM mode or the SPM mode into better play, and lays a mechanism foundation for enhancing performance of accessing the shared virtual memory.
  • FIG. 6 is a schematic structural diagram of a computer system according to the present disclosure.
  • the computer system in this embodiment includes at least one bus 601 , at least one processor 602 connected to the bus 601 , and at least one memory 603 connected to the bus 601 .
  • the processor 602 uses the bus 601 to call codes in the memory 603 to: monitor the access characteristics of the shared virtual memory page; and change the management mode of the shared virtual memory page to an SPM mode if the monitored access characteristics meet the first set condition and the current management mode of the shared virtual memory page is a DSM mode, and/or, change the management mode of the shared virtual memory page to a DSM mode if the monitored access characteristics meet the second set condition and the current management mode of the shared virtual memory page is an SPM mode.
  • the processor 602 can monitor the access characteristics of access performed by N processes for the shared virtual memory page (for convenience of citation, called a first shared virtual memory page below), where N is a positive integer.
  • N for convenience of citation, called a first shared virtual memory page below
  • N is a positive integer.
  • the N process may refer to any process currently existent in the system or a specific process; if N is greater than 1, the N processes may refer to part or all of the processes currently existent in the system.
  • the N processes may refer to part or all of the processes, which currently exist in the system and will possibly access the first shared virtual memory page.
  • a performance counter or other hardware or programs may be used to monitor the access characteristics of the shared virtual memory page.
  • the processor 602 can monitor various access characteristics of access performed by the process for the shared virtual memory page.
  • the access characteristics may include access frequency and/or access mode, and so on.
  • the access modes may include read-only access, read-write access, write access, and so on. That is, the processor 602 can monitor the access characteristics such as access frequency and/or access mode of access performed by the process for the shared virtual memory page.
  • the first set condition and the second set condition may vary with the monitored access characteristics. For example, if the monitored access characteristics are an access frequency, the first set condition and the second set condition may include set conditions related to the access frequency; if the monitored access characteristics are an access mode, the first set condition and the second set condition may include set conditions related to the access mode; if the monitored access characteristics include the access frequency and the access mode, the first set condition and the second set condition may include set conditions related to the both access frequency and the access mode.
  • the first set condition may include that the access frequency is less than or equal to a first threshold (for example, the first threshold may be 1000 times per second, 2000 times per second, 5000 times per second, or another frequency value); and the second set condition may include that the access frequency is greater than a second threshold, where the first threshold may be less than or equal to the second threshold, and the second threshold may be 1000 times per second, 2000 times per second, 3000 times per second, 5000 times per second, or another frequency value,
  • the management mode of the first shared virtual memory page may be changed to an SPM mode if the monitored access frequency is less than or equal to the first threshold and the current management mode of the first shared virtual memory page is a DSM mode, and/or, the management mode of the first shared virtual memory page may be changed to a DSM mode if the monitored access frequency is greater than the second threshold and the current management mode of the first shared virtual memory page is an
  • the first set condition may include that the access mode is write access or read-write access (or the percentage of the write access exceeds a fourth threshold or the frequency of the write access exceeds a third threshold); and the second set condition may include that the percentage of the read-only access or read access exceeds a sixth threshold or that the frequency of the read access exceeds a fifth threshold, where the sum of the fourth threshold and the sixth threshold may be greater than or equal to 1.
  • the processor 602 may change the management mode of the first shared virtual memory page to an SPM mode if the access mode monitored by the processor 602 is write access or read-write access (or the percentage of the write access exceeds a fourth threshold or the frequency of the write access exceeds a third threshold) and the current management mode of the first shared virtual memory page is a DSM mode; and/or, the processor 602 may change the management mode of the first shared virtual memory page to a DSM mode if the access mode monitored by the processor 602 is read-only access (or the percentage of the read access exceeds the sixth threshold or the frequency of the read access exceeds the fifth threshold) and the current management mode of the first shared virtual memory page is an SPM mode, and other circumstances may be inferred similarly.
  • the changing, by the processor 602 , the management mode of the first shared virtual memory page to a DSM mode may include: in a first physical memory space, allocating a first physical memory page to the first shared virtual memory page; copying data stored on a second physical memory page to the first physical memory page, where the second physical memory page belongs to a shared physical memory space (the shared physical memory space is a physical memory space that can be shared by several cores), and the second physical memory page is a physical memory page currently mapped from the first shared virtual memory page; creating a mapping relationship between the first shared virtual memory page and the first physical memory page; and recording the DSM mode as the management mode of the first shared virtual memory page.
  • the changing, by the processor 602 , the management mode of the shared virtual memory page to an SPM mode may include: creating a mapping relationship between the first shared virtual memory page and the third physical memory page, where the third physical memory page belongs to the shared physical memory space (the shared physical memory space is a physical memory space that can be shared by several cores); and recording the SPM mode as the management mode of the first shared virtual memory page, where the data stored on the physical memory page previously mapped from the shared virtual memory page remains unchanged.
  • the changing, by the processor 602 , the management mode of the shared virtual memory page to an SPM mode may include: synchronizing, by the processor 602 , changed data in the data stored on a fourth physical memory page to a third physical memory page; creating a mapping relationship between the first shared virtual memory page and the third physical memory page, where the third physical memory page belongs to the shared physical memory space (the shared physical memory space is a physical memory space that can be shared by several cores); and recording the SPM mode as the management mode of the first shared virtual memory page, where the fourth physical memory page is a physical memory page previously mapped from the shared virtual memory page and the data stored on the physical memory page is changed.
  • the processor 602 may record the DSM mode/SPM mode as the management mode of the first shared virtual memory page in a variety of ways. For example, a management mode flag bit is set for the first shared virtual memory page, and the management mode flag bit may be set to different values to identify whether the management mode of the first shared virtual memory page is a DSM mode or an SPM mode. Alternatively, the processor 602 may record the management mode of the first shared virtual memory page, which is recorded in the page information structure body of a process (or multiple processes or all processes), as a DSM mode or an SPM mode. Other ways of recording are not enumerated here exhaustively.
  • the processor 602 may set the management mode (such as DSM mode or SPM mode) for each shared virtual memory page (such as the first shared virtual memory page).
  • the management mode of the first shared virtual memory page is the same.
  • the management mode (such as DSM mode or SPM mode) corresponding to each process may also be set for each shared virtual memory page (such as the first shared virtual memory page).
  • the management mode of the first shared virtual memory page is the same or different. For example, assuming that process A and process B exist, the management mode corresponding to process A on the first shared virtual memory page may be a DSM mode, and the management mode corresponding to process B on the first shared virtual memory page may be an SPM mode.
  • the processor 602 may monitor access characteristics of access performed by the first process for the first shared virtual memory page, where the first process includes M 1 processes and M 1 is a positive integer; and the processor 602 changes the management mode corresponding to the first process on the first shared virtual memory page to an SPM mode if the monitored access characteristics meet the first set condition and the current management mode corresponding to the first process on the first shared virtual memory page is a DSM mode; and/or, the processor 602 may change the management mode corresponding to the first process on the first shared virtual memory page to a DSM mode if the monitored access characteristics meet the second set condition and the current management mode corresponding to the first process on the first shared virtual memory page is an SPM mode.
  • the processor 602 may further monitor the access characteristics of access performed by the second process for the first shared virtual memory page, where the second process includes M 2 processes, M 2 is a positive integer, and the second process is different from the first process; the processor 602 may change the management mode corresponding to the second process on the first shared virtual memory page to an SPM mode if the monitored access characteristics of access performed by the second process for the first shared virtual memory page meet the third set condition and the current management mode corresponding to the second process on the first shared virtual memory page is a DSM mode; and/or, the processor 602 may change the management mode corresponding to the second process on the first shared virtual memory page to a DSM mode if the monitored access characteristics of access performed by the second process for the first shared virtual memory page meet the fourth set condition and the current management mode corresponding to the second process on the first shared virtual memory page is an SPM mode, where the third set condition is the same as or different from the first condition, and the fourth set condition is the same as or different from the second condition.
  • the processor 602 may use an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored for each process in the first process (or the second process) that accesses the shared virtual memory page, as the monitored access characteristics of the first process (or the second process) that accesses the shared virtual memory page; or, use the access characteristics, which are monitored for one process in the first process (or the second process) that accesses the shared virtual memory page, as the monitored access characteristics of the first process (or the second process) that accesses the shared virtual memory page; and change the management mode corresponding to the first process (or the second process) on the shared virtual memory page to a shared physical memory mode if the monitored access characteristics of access performed by the processes for the shared virtual memory page meet the first condition and if the current management mode corresponding to the first process (or the second process) on the shared virtual memory page is a distributed shared memory mode.
  • the processor 602 may use an average, a weighted average, a sum, or a weighted sum of access characteristics, which are respectively monitored for each process in the first process (or the second process) that accesses the shared virtual memory page, as the monitored access characteristics of the first process (or the second process) that accesses the shared virtual memory page; or, use the access characteristics, which are monitored for one process in the first process (or the second process) that accesses the shared virtual memory page, as the monitored access characteristics of the first process (or the second process) that accesses the shared virtual memory page; and change the management mode corresponding to the first process (or the second process) on the shared virtual memory page to a distributed shared memory mode if the monitored access characteristics of access performed by the processes for the shared virtual memory page meet the second condition and if the current management mode corresponding to the first process (or the second process) on the shared virtual memory page is a shared physical memory mode.
  • FIG. 7 is a schematic diagram of another computer system 600 according to an embodiment of the present disclosure.
  • the computer system 600 further includes an input apparatus 604 and an output apparatus 605 , where the input apparatus 604 and the output apparatus 605 are connected to a bus 601 .
  • FIG. 6 or FIG. 7 is only a schematic structural diagram of a computer system provided in the present disclosure, and the specific structure may be adjusted according to actual conditions.
  • the computer system 600 in this embodiment may be a communication device such as an access network device or a core network device or a router or another communication device, and the computer system 600 may be a server or a personal computer or another computer device.
  • An embodiment of the present disclosure further provides a computer storage medium.
  • the computer storage medium may store programs. When the programs run, the programs can execute part or all of the steps of the method for determining a management mode of a shared virtual memory page as described in the method embodiment.
  • the disclosed apparatus may be implemented in other manners.
  • the described apparatus embodiment is merely exemplary.
  • the unit division is merely logical function division and there may be other division in actual implementation.
  • a plurality of units or components may be combined or integrated into another system, or some features may be ignored or not performed.
  • the displayed or discussed mutual couplings, direct couplings, or communication connections may be implemented through some interfaces.
  • the indirect couplings or communication connections between the apparatuses or units may be implemented in electronic, mechanical, or other forms.
  • the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one position, or may also be distributed on a plurality of network units. Some or all of the units may be selected to achieve the objective of the solution of the embodiment according to an actual need.
  • functional units in the embodiments of the present disclosure may be integrated into one processing unit, or each of the units may also exist alone physically, or two or more units may also be integrated into one unit.
  • the integrated units may be implemented in a form of hardware, or may also be implemented in a form of a software functional unit.
  • the integrated units When the integrated units are implemented in a form of a software functional unit and sold or used as an independent product, the integrated units may be stored in a computer-readable storage medium.
  • the computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which may be a personal computer, a server, a network device, or the like) to perform all or a part of the steps of the methods described in the embodiments of the present disclosure.
  • the foregoing storage medium includes: any medium that can store program codes, such as a USB flash disk, a read-only memory (ROM), a random access memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Human Computer Interaction (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Computer Security & Cryptography (AREA)
  • Memory System Of A Hierarchy Structure (AREA)
  • Multi Processors (AREA)
  • Mathematical Physics (AREA)
US14/749,869 2012-12-25 2015-06-25 Method and related device for determining management mode of shared virtual memory page Active 2035-03-10 US10019178B2 (en)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN201210572056 2012-12-25
CN201210572056.5 2012-12-25
CN201210572056.5A CN103902459B (zh) 2012-12-25 2012-12-25 确定共享虚拟内存页面管理模式的方法和相关设备
PCT/CN2013/080863 WO2014101423A1 (zh) 2012-12-25 2013-08-06 确定共享虚拟内存页面管理模式的方法和相关设备

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/080863 Continuation WO2014101423A1 (zh) 2012-12-25 2013-08-06 确定共享虚拟内存页面管理模式的方法和相关设备

Publications (2)

Publication Number Publication Date
US20150293720A1 US20150293720A1 (en) 2015-10-15
US10019178B2 true US10019178B2 (en) 2018-07-10

Family

ID=50993794

Family Applications (1)

Application Number Title Priority Date Filing Date
US14/749,869 Active 2035-03-10 US10019178B2 (en) 2012-12-25 2015-06-25 Method and related device for determining management mode of shared virtual memory page

Country Status (9)

Country Link
US (1) US10019178B2 (de)
EP (1) EP2940595B1 (de)
JP (1) JP6296304B2 (de)
KR (1) KR101698950B1 (de)
CN (2) CN107402891B (de)
AU (1) AU2013370724B2 (de)
MX (1) MX346986B (de)
RU (1) RU2607984C1 (de)
WO (1) WO2014101423A1 (de)

Families Citing this family (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10095582B2 (en) * 2014-10-29 2018-10-09 International Business Machines Corporation Partial rebuilding techniques in a dispersed storage unit
US20180181332A1 (en) * 2014-10-29 2018-06-28 International Business Machines Corporation Expanding a dispersed storage network memory beyond two locations
US20180113747A1 (en) * 2014-10-29 2018-04-26 International Business Machines Corporation Overdrive mode for distributed storage networks
US10459792B2 (en) * 2014-10-29 2019-10-29 Pure Storage, Inc. Using an eventually consistent dispersed memory to implement storage tiers
CN104486415B (zh) * 2014-12-15 2018-06-15 新浪网技术(中国)有限公司 监控对象工作状态的判定方法及装置
US10019373B2 (en) * 2014-12-19 2018-07-10 Mediatek Inc. Memory management method for supporting shared virtual memories with hybrid page table utilization and related machine readable medium
US10228990B2 (en) * 2015-11-12 2019-03-12 Sandisk Technologies Llc Variable-term error metrics adjustment
CN105426322B (zh) * 2015-12-31 2019-05-28 华为技术有限公司 一种数据的预取方法及装置
US10732856B2 (en) 2016-03-03 2020-08-04 Sandisk Technologies Llc Erase health metric to rank memory portions
CN105808322B (zh) * 2016-04-01 2019-05-03 深圳市硅格半导体股份有限公司 多分区虚拟内存的映射控制方法及装置
US10055155B2 (en) * 2016-05-27 2018-08-21 Wind River Systems, Inc. Secure system on chip
WO2019105565A1 (en) * 2017-12-01 2019-06-06 Huawei Technologies Co., Ltd. Systems for compiling and executing code within one or more virtual memory pages
CN109240602B (zh) * 2018-08-06 2021-01-15 联想(北京)有限公司 数据存取方法
CN109408202B (zh) * 2018-09-28 2020-11-06 新华三技术有限公司 一种lmdb扩展方法及装置
CN109582592B (zh) * 2018-10-26 2021-06-15 华为技术有限公司 资源管理的方法和装置
KR102144011B1 (ko) * 2019-02-15 2020-08-12 성균관대학교산학협력단 가상화 서비스에서 페이지 캐시 관리 방법 및 장치
CN110083387B (zh) 2019-04-19 2021-11-12 上海兆芯集成电路有限公司 使用轮询机制的处理系统及其访存方法
CN110058931B (zh) 2019-04-19 2022-03-22 上海兆芯集成电路有限公司 用以任务调度的处理系统及其加速方法
CN110032453B (zh) 2019-04-19 2022-05-03 上海兆芯集成电路有限公司 用以任务调度与分配的处理系统及其加速方法
CN110046053B (zh) * 2019-04-19 2021-11-12 上海兆芯集成电路有限公司 用以分配任务的处理系统及其访存方法
CN110618883B (zh) * 2019-09-26 2022-09-13 迈普通信技术股份有限公司 一种用于共享内存链表的方法、装置、设备及存储介质
US11972034B1 (en) 2020-10-29 2024-04-30 Amazon Technologies, Inc. Hardware-assisted obscuring of cache access patterns
CN116107742A (zh) * 2021-06-10 2023-05-12 荣耀终端有限公司 虚拟内存管理方法和电子设备
US11635919B1 (en) * 2021-09-30 2023-04-25 Amazon Technologies, Inc. Safe sharing of hot and cold memory pages
CN114528123B (zh) * 2022-02-11 2024-12-06 北京天融信网络安全技术有限公司 数据访问方法、装置、设备及计算机可读存储介质
CN114661497B (zh) * 2022-03-31 2023-01-10 慧之安信息技术股份有限公司 操作系统分区共享内存方法和系统
CN115659077B (zh) * 2022-09-09 2025-09-26 福建天晴在线互动科技有限公司 一种内存优化方法及其系统
CN117112268B (zh) * 2023-10-23 2024-02-13 深圳市七彩虹禹贡科技发展有限公司 一种内存共享管理方法及系统

Citations (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5630097A (en) 1991-06-17 1997-05-13 Digital Equipment Corporation Enhanced cache operation with remapping of pages for optimizing data relocation from addresses causing cache misses
CN1149736A (zh) 1995-06-05 1997-05-14 北京航空航天大学 一种分布式存储器共享管理方法
EP0780770A1 (de) 1995-12-22 1997-06-25 Sun Microsystems, Inc. Hybrides NUMA/COMA-Cachespeicherungssystem und Verfahren zur Auswahl aus den Cachemoden
EP0848330A2 (de) 1996-12-11 1998-06-17 Ncr International Inc. Lagesteuerung von Speicherseiten für Mehrspeicher-Mehrprozessor-System
US5875487A (en) 1995-06-07 1999-02-23 International Business Machines Corporation System and method for providing efficient shared memory in a virtual memory system
US6075938A (en) 1997-06-10 2000-06-13 The Board Of Trustees Of The Leland Stanford Junior University Virtual machine monitors for scalable multiprocessors
JP2001175619A (ja) 1999-12-22 2001-06-29 Univ Waseda シングルチップマルチプロセッサ
US20020013889A1 (en) 1998-09-28 2002-01-31 Assaf Schuster Distributed shared memory system with variable granularity
US20030061395A1 (en) * 1998-06-29 2003-03-27 Brent A. Kingsbury Message passing using shared memory of a computer
US20040073743A1 (en) 2002-10-10 2004-04-15 International Business Machines Corp. Method and system of managing virtualized physical memory in a multi-processor system
US6910106B2 (en) 2002-10-04 2005-06-21 Microsoft Corporation Methods and mechanisms for proactive memory management
US20070011420A1 (en) 2005-07-05 2007-01-11 Boss Gregory J Systems and methods for memory migration
US20070180448A1 (en) 2006-01-24 2007-08-02 Citrix Systems, Inc. Methods and systems for providing access to a computing environment provided by a virtual machine executing in a hypervisor executing in a terminal services session
CN101398768A (zh) 2008-10-28 2009-04-01 北京航空航天大学 一种分布式虚拟机监视器系统的构建方法
CN101477496A (zh) 2008-12-29 2009-07-08 北京航空航天大学 基于分布式内存虚拟化的numa结构的实现方法
JP2011065650A (ja) 2009-09-18 2011-03-31 Intel Corp ローカル物理メモリとリモート物理メモリとの間で共有されるバーチャルメモリのためのハードウェアサポートの提供
US8135936B2 (en) 2009-12-23 2012-03-13 Intel Corporation Adaptive address mapping with dynamic runtime memory mapping selection
US8156305B1 (en) 2008-10-01 2012-04-10 Netapp, Inc. Remapping of data addresses for large capacity low-latency random read memory
US8195914B2 (en) 2002-12-27 2012-06-05 Intel Corporation Mechanism for remapping post virtual machine memory pages
US9208088B2 (en) * 2012-01-03 2015-12-08 Seoul National University R&Db Foundation Shared virtual memory management apparatus for providing cache-coherence
US20160034397A1 (en) * 2013-04-16 2016-02-04 Huawei Technologies Co., Ltd. Method and Apparatus for Processing Data and Computer System

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006065525A (ja) * 2004-08-26 2006-03-09 Hitachi Ltd 分散環境における仮想共有空間の実現方法
CN101477495B (zh) * 2008-10-28 2011-03-16 北京航空航天大学 分布式内存虚拟化技术的实现方法

Patent Citations (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5630097A (en) 1991-06-17 1997-05-13 Digital Equipment Corporation Enhanced cache operation with remapping of pages for optimizing data relocation from addresses causing cache misses
CN1149736A (zh) 1995-06-05 1997-05-14 北京航空航天大学 一种分布式存储器共享管理方法
US5875487A (en) 1995-06-07 1999-02-23 International Business Machines Corporation System and method for providing efficient shared memory in a virtual memory system
EP0780770A1 (de) 1995-12-22 1997-06-25 Sun Microsystems, Inc. Hybrides NUMA/COMA-Cachespeicherungssystem und Verfahren zur Auswahl aus den Cachemoden
JPH09185550A (ja) 1995-12-22 1997-07-15 Sun Microsyst Inc キャッシュ記憶モードの間で選択するためのハイブリッドｎｕｍａ ｃｏｍａキャッシュ記憶のシステムおよび方法
EP0848330A2 (de) 1996-12-11 1998-06-17 Ncr International Inc. Lagesteuerung von Speicherseiten für Mehrspeicher-Mehrprozessor-System
JPH10320272A (ja) 1996-12-11 1998-12-04 Ncr Internatl Inc 多重プロセッサを有するコンピュータ・システム及びそのためのメモリ・ページ位置制御方法
US6075938A (en) 1997-06-10 2000-06-13 The Board Of Trustees Of The Leland Stanford Junior University Virtual machine monitors for scalable multiprocessors
US20030061395A1 (en) * 1998-06-29 2003-03-27 Brent A. Kingsbury Message passing using shared memory of a computer
US20020013889A1 (en) 1998-09-28 2002-01-31 Assaf Schuster Distributed shared memory system with variable granularity
JP2001175619A (ja) 1999-12-22 2001-06-29 Univ Waseda シングルチップマルチプロセッサ
US6910106B2 (en) 2002-10-04 2005-06-21 Microsoft Corporation Methods and mechanisms for proactive memory management
RU2348067C2 (ru) 2002-10-04 2009-02-27 Майкрософт Корпорейшн Способы и устройства для упреждающего управления памятью
US20040073743A1 (en) 2002-10-10 2004-04-15 International Business Machines Corp. Method and system of managing virtualized physical memory in a multi-processor system
KR20040032745A (ko) 2002-10-10 2004-04-17 인터내셔널 비지네스 머신즈 코포레이션 멀티프로세서 시스템에서 가상화된 물리적 메모리를관리하는 방법 및 시스템
US8195914B2 (en) 2002-12-27 2012-06-05 Intel Corporation Mechanism for remapping post virtual machine memory pages
US20070011420A1 (en) 2005-07-05 2007-01-11 Boss Gregory J Systems and methods for memory migration
JP2009500705A (ja) 2005-07-05 2009-01-08 インターナショナル・ビジネス・マシーンズ・コーポレーション メモリ・マイグレーションのシステムおよび方法
WO2007003630A1 (en) 2005-07-05 2007-01-11 International Business Machines Corporation Systems and methods for memory migration
US20070180448A1 (en) 2006-01-24 2007-08-02 Citrix Systems, Inc. Methods and systems for providing access to a computing environment provided by a virtual machine executing in a hypervisor executing in a terminal services session
US8156305B1 (en) 2008-10-01 2012-04-10 Netapp, Inc. Remapping of data addresses for large capacity low-latency random read memory
CN101398768A (zh) 2008-10-28 2009-04-01 北京航空航天大学 一种分布式虚拟机监视器系统的构建方法
CN101477496A (zh) 2008-12-29 2009-07-08 北京航空航天大学 基于分布式内存虚拟化的numa结构的实现方法
JP2011065650A (ja) 2009-09-18 2011-03-31 Intel Corp ローカル物理メモリとリモート物理メモリとの間で共有されるバーチャルメモリのためのハードウェアサポートの提供
US8135936B2 (en) 2009-12-23 2012-03-13 Intel Corporation Adaptive address mapping with dynamic runtime memory mapping selection
US9208088B2 (en) * 2012-01-03 2015-12-08 Seoul National University R&Db Foundation Shared virtual memory management apparatus for providing cache-coherence
US20160034397A1 (en) * 2013-04-16 2016-02-04 Huawei Technologies Co., Ltd. Method and Apparatus for Processing Data and Computer System
US9824011B2 (en) * 2013-04-16 2017-11-21 Huawei Technologies Co., Ltd. Method and apparatus for processing data and computer system

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
Junghyun Kim et al: "An Efficient Software Shared Virtual Memory for the Single-chip Cloud Computer", 2011, total 5 pages.
Osamu Tatebe, et al., "OpenMP compiler for PC clusters using TDL distributed array library", Inforamtion Processing Society of Japan Techncial Reports, Information Processing Society of Japan, Jul. 27, 2001, vol. 2001, No. 77 (2001-HPC-87), pp. 123-128, with English abctract.
Stefan Lankes et al: "The Path to Metal SVM: Shared Virtual Memory for the SCC", Dec. 2011, total 8 pages.
Xiaocheng Zhou et al: "A Case for Software Managed Coherence in Many-core Processors", Intel Corporation, 2010, total 6 pages.
Yukio Oishi, et al., "Two-level Scheduling in Collaboration with Memory Management for NUMA Multiprocessors", Inforamtion Processing Society of Japan Techncial Reports, Information Processing Society of Japan, Aug. 23, 1995, vol. 95, No. 79 (95-OS-70), pp. 97-104. with English abctract.

Also Published As

Publication number Publication date
CN103902459A (zh) 2014-07-02
EP2940595B1 (de) 2020-02-12
MX346986B (es) 2017-04-07
CN103902459B (zh) 2017-07-28
US20150293720A1 (en) 2015-10-15
EP2940595A1 (de) 2015-11-04
JP2016502213A (ja) 2016-01-21
AU2013370724A1 (en) 2015-07-09
MX2015008281A (es) 2015-09-29
WO2014101423A1 (zh) 2014-07-03
AU2013370724B2 (en) 2017-01-12
KR20150091158A (ko) 2015-08-07
KR101698950B1 (ko) 2017-02-01
RU2607984C1 (ru) 2017-01-11
EP2940595A4 (de) 2016-01-13
CN107402891B (zh) 2020-12-22
CN107402891A (zh) 2017-11-28
JP6296304B2 (ja) 2018-03-20

Similar Documents

Publication Publication Date Title
US20150293720A1 (en) Method and related device for determining management mode of shared virtual memory page
US9824011B2 (en) Method and apparatus for processing data and computer system
KR102380670B1 (ko) 메모리 제어기에서의 세분화된 대역폭 프로비저닝
US10496613B2 (en) Method for processing input/output request, host, server, and virtual machine
CN104636187B (zh) 基于负载预测的numa架构中虚拟机调度方法
CN113590508B (zh) 动态可重构的内存地址映射方法及装置
CN104461735B (zh) 一种虚拟化场景下分配cpu资源的方法和装置
WO2016019566A1 (zh) 内存管理方法、装置和系统、以及片上网络
TWI871157B (zh) 用於記憶體管理機會與記憶體交換任務之運算系統及管理其之方法
JP2023527770A (ja) メモリにおける推論
CN114020416A (zh) 大页内存动态管理方法、装置和计算机设备
CN105426324B (zh) 终端设备的内存访问控制方法与装置
CN109412865B (zh) 一种虚拟网络资源分配方法、系统及电子设备
CN104281545B (zh) 一种数据读取方法及设备
CN113678104B (zh) 设备内符号数据移动系统
US20180173639A1 (en) Memory access method, apparatus, and system
CN105988871B (zh) 一种远端内存分配方法、装置和系统
CN104281532A (zh) 一种基于numa架构的虚拟机内存访问监测方法
Kishani et al. Padsa: Priority-aware block data storage architecture for edge cloud serving autonomous vehicles
CN107273188B (zh) 一种虚拟机中央处理单元cpu绑定方法及装置
US20130111177A1 (en) Implementing feedback directed numa mitigation tuning
KR102964796B1 (ko) 컴퓨팅 시스템에서의 메모리 티어링 기법
CN106155939B (zh) 一种信息处理方法及电子设备
CN121433810A (zh) 一种多层级自匹配的虚拟机内存分配方法及系统
CN115604269A (zh) 一种服务器的负载均衡方法、装置、电子设备及存储介质

Legal Events

Date Code Title Description
AS Assignment

Owner name: HUAWEI TECHNOLOGIES CO., LTD., CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:LAM, KINGTIN;ZHU, WANGBIN;YAN, YOULIANG;AND OTHERS;SIGNING DATES FROM 20150615 TO 20150617;REEL/FRAME:035965/0918

STCF Information on status: patent grant

Free format text: PATENTED CASE

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 4TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1551); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 4

MAFP Maintenance fee payment

Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1552); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

Year of fee payment: 8